A non-invasive combination of liver stiffness measurements and blood tests could substantially improve the early detection of sinusoidal obstruction syndrome, a dangerous complication of allogeneic hematopoietic cell transplantation, according to a prospective study of 180 transplant recipients. The research, published in Annals of Hematology, found that transient elastography—a technology that estimates tissue stiffness using mechanically induced waves—performed particularly well when combined with measurements of plasminogen activator inhibitor-1 (PAI-1) and type III procollagen peptide (P3P). For patients who developed liver injury after transplantation, the combined markers produced an area under the receiver operating characteristic curve of 0.96, indicating a strong ability to distinguish SOS from other causes of post-transplant liver damage. The findings suggest that a bedside assessment integrating imaging and molecular signals could help clinicians identify patients before the syndrome becomes irreversible.
Sinusoidal obstruction syndrome, also known as hepatic veno-occlusive disease, arises when the smallest blood vessels within the liver become damaged and narrowed. These vessels, called hepatic sinusoids, form a delicate interface between circulating blood and liver cells. Conditioning treatments used before transplantation, including chemotherapy and sometimes radiation, can injure the sinusoidal lining and the surrounding extracellular matrix. Swollen endothelial cells, disrupted vessel walls and debris can progressively obstruct blood flow through the liver. As pressure builds upstream, the liver enlarges, fluid accumulates in the abdomen and body weight rises. In severe cases, impaired circulation can lead to jaundice, kidney dysfunction, respiratory failure and death. Because the syndrome may progress rapidly, recognizing it during its earliest phase is crucial for treatment and for decisions about reducing additional liver stress.
Clinicians traditionally rely on the Seattle and Baltimore diagnostic criteria, which use combinations of weight gain, painful liver enlargement, jaundice, fluid retention and the timing of symptoms after transplantation. These criteria remain useful, but they can be difficult to apply in patients whose symptoms overlap with infection, drug toxicity, graft-versus-host disease or fluid overload. Their sensitivity and positive predictive value can also be limited, particularly when SOS develops without the classic presentation. A liver biopsy can provide direct pathological evidence of sinusoidal injury and venous obstruction, but performing the procedure in a patient with low platelet counts or impaired clotting carries a potentially serious risk of hemorrhage. The Japanese research team therefore designed its study around tools that could be repeated without penetrating the liver: abdominal ultrasonography, transient elastography and blood-based biomarkers associated with vascular injury and tissue remodeling.
The investigators prospectively followed 180 people undergoing allogeneic hematopoietic cell transplantation. In this procedure, patients receive blood-forming stem cells from a donor after conditioning therapy has suppressed or eliminated their own marrow. The treatment is potentially curative for several blood cancers and other hematological disorders, but it places multiple organs under considerable physiological stress. Within the study cohort, 10 patients developed definitive SOS. Most diagnoses were supported by pathological examination of liver tissue obtained by biopsy or autopsy, giving the researchers a stringent reference standard rather than relying only on clinical suspicion. The team also focused on a subgroup of 48 patients who developed liver injury after transplantation, a clinically challenging population in which the symptoms of SOS can be confused with more common transplant-related complications.
Transient elastography measures how rapidly a shear wave travels through tissue. A probe placed against the body generates a brief mechanical vibration, and ultrasound tracks the resulting wave as it moves through the liver. Stiffer tissue transmits the wave more rapidly, allowing the instrument to calculate a liver-stiffness value. Fibrosis is a familiar cause of elevated stiffness, but in acute SOS the measurement may change for a different reason: congestion and increased pressure within the hepatic microcirculation can make the liver mechanically harder even before extensive scarring develops. This distinction makes serial measurements potentially informative. In the study, liver stiffness emerged as one of the strongest indicators of SOS. Among patients with post-transplant liver injury, its maximum AUC was 0.83, with a 95 percent confidence interval of 0.67 to 0.98. The test is rapid and non-invasive, although inflammation, cholestasis, congestion and technical factors can also influence stiffness readings.
The strongest blood-based diagnostic signal came from PAI-1, a protein that suppresses fibrinolysis, the process by which the body dissolves blood clots. PAI-1 is produced by endothelial and other cells and can rise when the vascular lining is activated or damaged. In SOS, injury to sinusoidal endothelial cells may promote a pro-thrombotic state and reduce the ability of blood vessels to clear fibrin, contributing to microvascular obstruction. PAI-1 therefore offers a window into the biological process rather than simply reflecting its consequences. In the liver-injury subgroup, PAI-1 reached an AUC of 0.85, with a 95 percent confidence interval of 0.67 to 1.00. When PAI-1 was combined with liver stiffness, diagnostic performance improved to an AUC of 0.96, suggesting that the two measurements capture complementary dimensions of disease: a biochemical signal of endothelial dysfunction and a physical signal of altered liver mechanics.
The study also identified a role for P3P in predicting SOS before transplantation. Type III procollagen peptide is released during the production of type III collagen, a structural component of connective tissue and the extracellular matrix. When procollagen is synthesized, portions of the precursor molecule are cleaved and released into the bloodstream, making P3P a marker of active matrix formation or remodeling. Sinusoidal injury can disturb the space surrounding the hepatic microvasculature and stimulate extracellular-matrix changes, potentially explaining why elevated pre-transplant P3P was associated with later SOS. Across all 180 patients, pre-transplant P3P produced an AUC of 0.82, while pre-transplant liver stiffness reached 0.80. These values indicate useful discrimination in this cohort, though they do not mean that either test alone can establish a diagnosis or predict every future case with certainty.
Perhaps the most revealing evidence came from the researchers’ kinetic analysis, which examined how the markers changed over time rather than treating each measurement as an isolated number. Liver stiffness, PAI-1 and P3P rose during the early period after transplantation specifically in patients who went on to develop SOS. A single elevated result can be difficult to interpret because transplant recipients frequently experience inflammation, medication effects and temporary changes in fluid balance. A rising trajectory, by contrast, may signal that the underlying vascular injury is progressing. This temporal pattern supports a monitoring strategy in which patients receive repeated assessments before and after transplantation. A sudden increase in stiffness accompanied by rising PAI-1 could prompt closer fluid management, review of potentially hepatotoxic medicines and consideration of disease-directed therapy, while P3P could help identify higher-risk patients even before conditioning begins.
Abdominal ultrasonography was included to assess conventional structural and haemodynamic features of liver disease, such as liver enlargement, ascites, gallbladder-wall changes and alterations in portal blood flow. Ultrasound is widely available and avoids radiation, but its diagnostic value can depend heavily on operator experience, patient anatomy and the stage of disease. The study’s broader contribution was not to replace ultrasound, but to place it alongside elastography and biomarkers in the same prospective cohort. That comparison is important because previous reports often evaluated individual tests in different patient populations, using varying definitions of SOS and different timing schedules. By applying a pathological definition wherever possible, the investigators sought to reduce uncertainty about what the tests were actually detecting. The results point toward a multimodal approach rather than a single “magic” measurement, with imaging and blood assays potentially compensating for one another’s blind spots.
The findings could influence how transplant teams screen for SOS, but they do not yet establish a universal diagnostic algorithm. Only 10 of the 180 participants developed definitive disease, so estimates of test performance are based on a relatively small number of cases and include wide confidence intervals. The cohort was drawn from institutions in Japan, and the results may not automatically apply to patients receiving different conditioning regimens, transplant protocols or supportive care. Biomarker thresholds, assay standardization and the frequency of elastography measurements will also need to be defined before routine implementation. Even so, the study offers a compelling framework for earlier, safer surveillance. By combining P3P to assess baseline susceptibility, PAI-1 to detect endothelial and coagulation-related injury, and liver stiffness to track mechanical consequences within the organ, clinicians may gain a more detailed picture of SOS as it develops. For a complication in which diagnostic delay can be life-threatening and biopsy can be hazardous, that integrated view could become a valuable step toward more precise transplant medicine.
